Value of Analytics is in Consumption - Operationalize it

Many large organization’s today have embarked on a journey of Digital Transformation.

One of the key tenets within it is to adopt and drive Analytics and Insights based Decisions across the Enterprise. Naturally, we’ve seen companies investing a great deal of time and money to make this a reality.

In 2019 alone, IDC estimates that organizations invested $189.1 billion in analytics. By 2022, they predict spending will increase to $274.3 billion.

However, despite such significant investments:

  • Less than half of the models that are developed get deployed.
  • 90% of models take more than three months to deploy.
  • Nearly 50% take over seven months to be put into production.

Deployment, the “last mile” of analytics, remains a formidable challenge.

Having said that some of the world’s leading technology firms that specialize in Analytics and Decisioning, are investing heavily in their R&D to build unified platforms that will manage the complete Analytics Life Cycle.

Coupled with synergizing Data Access & Data Governance Technologies and related business processes and workflows, these companies are helping address exactly the challenge stated above.

So today, friends, I will primarily talk about the “Last Mile” challenge and also discuss how SAS has been helping address this problem, for the developer and data science community at large, both Open Source and SAS.

I will walk you through the full Analytics Life Cycle (buildà compareà publishàgovernà consume), and time permitting showcase a small demonstration of the same.

 
 

Outline/Structure of the Demonstration

  • PPT
    • Life cycle of machine learning
    • Challenges of deployment
    • How SAS can help
  • Demo

Learning Outcome

Audience would see the complete analytical life cycle from building a model to deployment to monitor the performance of the model. The audience would also see how a complete model deployment life cycle could be automated and how it can collaborate multiple teams.

Target Audience

Data scientist, IT teams

schedule Submitted 1 year ago

Public Feedback

    help